NVIDIA RTX 50 SUPER Refresh Delayed to CES With Major VRAM Upgrades and Future RTX 60 Rubin Architecture Launch Timeline Revealed
The ongoing global buildout of AI infrastructure has created a huge shortage in high speed memory components, directly impacting consumer graphics card market. Initially NVIDIA planned to launch their mid generation GeForce RTX 50 SUPER refresh much earlier, but due to the memory shortage, they pushed it back for strategic reasons. Desktop gamers looking forward to get the updated hardware should expect to wait till early next year to see the official launch at the annual consumer trade show CES.
As revealed by leak details from Taiwanese site BenchLife.info, the main focus of the hardware refresh is a substantial jump in the video memory amount. NVIDIA is switching the board memory from 16Gb to 24Gb GDDR7 chips to pack the board with 50% more video memory. Following this configuration, the high end RTX 5080 SUPER and the RTX 5070 Ti SUPER both get 24 GB memory while their non SUPER counterparts pack 16 GB. Lowering down the stack, the RTX 5070 SUPER should feature 18 GB of memory instead of 12 GB and budget RTX 5060 SUPER get 12 GB vs 8 GB on the base model. For the moment, these changes seem to apply to desktop hardware, not for laptops.
If you are a gamer patiently waiting for a true generational jump under the Rubin architecture (likely named RTX 60 series), then you have to wait a bit longer. According to industry analysis, NVIDIA typically waits at least one year after a SUPER refresh has been available on market before introducing the entirely new architecture. Due to current generation delays, it is unlikely the gaming lineup under Rubin architecture to be launched until fall of next year.
The exact launch window of the RTX 60 series could easily shift back further depending on production yields and stabilization of the global memory supply chain. Rumors suggest the official launch would then happen again either during CES or GDC early next year. It seems the launch schedule for consumer gaming hardware increasingly is dominated by the huge demands for components in the enterprise AI market.
